REPLAY: Modeling Time-Varying Temporal Regularities of Human Mobility for Location Prediction over Sparse Trajectories

2024-02-26

IEEE Transactions on Mobile Computing

AI Summary

To address the challenge of modeling time-varying regularity in user location prediction from sparse trajectories, this paper proposes a novel RNN architecture that introduces, for the first time, Gaussian-weighted timestamp embeddings with learnable, time-varying bandwidths—explicitly capturing the continuous change in mobility regularity across periods (e.g., strong during morning rush hour vs. weak at night). The method jointly models spatiotemporal distance and dynamic temporal regularity, leveraging the adaptive bandwidth mechanism to capture human mobility’s time-varying periodicity and enabling effective retrieval of historical hidden states under sparsity. Evaluated on two real-world datasets, our approach outperforms state-of-the-art models by 7.7%–10.5% in prediction accuracy. Crucially, the learned bandwidth parameters exhibit clear spatiotemporal interpretability—quantitatively confirming well-known patterns, such as heightened regularity during rush hours and diminished regularity at night.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Machine Learning: Time-Series/Data Streams
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution

Abstract

Location prediction forecasts a user's location based on historical user mobility traces. To tackle the intrinsic sparsity issue of real-world user mobility traces, spatiotemporal contexts have been shown as significantly useful. Existing solutions mostly incorporate spatiotemporal distances between locations in mobility traces, either by feeding them as additional inputs to Recurrent Neural Networks (RNNs) or by using them to search for informative past hidden states for prediction. However, such distance-based methods fail to capture the time-varying temporal regularities of human mobility, where human mobility is often more regular in the morning than in other periods, for example; this suggests the usefulness of the actual timestamps besides the temporal distances. Against this background, we propose REPLAY, a general RNN architecture learning to capture the time-varying temporal regularities for location prediction. Specifically, REPLAY not only resorts to the spatiotemporal distances in sparse trajectories to search for the informative past hidden states, but also accommodates the time-varying temporal regularities by incorporating smoothed timestamp embeddings using Gaussian weighted averaging with timestamp-specific learnable bandwidths, which can flexibly adapt to the temporal regularities of different strengths across different timestamps. Our extensive evaluation compares REPLAY against a sizable collection of state-of-the-art techniques on two real-world datasets. Results show that REPLAY consistently and significantly outperforms state-of-the-art methods by 7.7%-10.5% in the location prediction task, and the bandwidths reveal interesting patterns of the time-varying temporal regularities.

Problem

• Predicting user locations from sparse mobility traces
• Capturing time-varying temporal regularities in human mobility
• Improving accuracy by incorporating timestamp embeddings and spatiotemporal distances

Innovation

• Uses RNN to capture time-varying mobility regularities
• Incorporates smoothed timestamp embeddings adaptively
• Employs Gaussian weighted averaging with learnable bandwidths